The term “valley of death” is a colorful (and biblical) reference to the difficulty of bringing scientific advances to the market. Researchers make a discovery in the lab that has a potential practical application. They then create a start up company to translate their discovery into a marketable product or service. The valley of death is the gulf between the lab and a profitable product, a desert that turns out to be too long for many, resulting in funding drying up before the market is reached.

As someone who is interested in science and technology, I have witnessed the valley of death many times from the sidelines. Often, when a scientist makes an interesting discovery, a science journalist reporting on the discovery feels obliged to connect the advance to some practical application. The more this application resembles technology from popular science fiction the better.

I enjoy speculating about future applications as much as anyone, but this practice can become formulaic and mindless. Every discovery about a virus will cure the common cold, every advance in understanding the machinery of cells will cure cancer, and every material science advance will give us hover cars or invisibility cloaks.

Another pattern that has emerged is the “5-10 years” claim, which is how long it will always take for the advance being reported to be translated to the marketplace. Often the scientists themselves are actively involved in the hype and overly optimistic predictions. Someone cynical might interpret “5-10 years” as “one more funding cycle.”

A recent commentary in Physics World on the valley of death offers some insight into this pattern of reporting. Scientists who have made a potentially useful discovery often do try to carry their discovery through the valley of death. They need funding, and often their project fails because funding runs out, and so a little hype might just provide the investors they need.

One problem is that it is remarkably difficult to translate discoveries to a workable product, and scientists don’t necessarily have the expertise to even see the difficulties ahead.

Battery technology is an excellent example of this. Almost every week I see a press release discussing the latest possible advance in battery technology that may revolutionize the batteries we use every day in 5-10 years. Somewhere deep in the article, maybe, we get a discussion of the hurdles. All they have to do is scale up the process, or figure out how to mass produce their technology. Often there are still fatal hurdles that need to be overcome.

For example, in order for a battery to be marketable it has to have a certain suite of features simultaneously: High capacity to both volume and weight, rapid recharge, sufficient discharge to power whatever device it’s being used for, a large number of recharge-discharge cycles, stability (it’s not likely to catch fire or explode), made from materials that are not overly expensive or toxic (complicating disposal), and able to be mass produced. Lacking any one of these features can be a fatal flaw. So we might hear about an amazing breakthrough that doubles capacity, and we only read deeper in that it requires prohibitive amounts of platinum, but the researchers are confident they can find a more affordable substitute.

Solar panels are another example. In both technologies we are making incremental advances, but nothing game-changing. For solar panels we want high efficiency, but also long life, flexibility would be nice, and cheap to mass produce. In the very same issue of physics world there is an article about a demonstration of a new plasmovoltaic effect. The researchers show by using a laser that plasmons on the surface of a metal can be made to produce a voltage difference. The article states:

While the devices reported by the team simply produce a potential difference when illuminated, the team is now working on a device that will deliver usable electrical energy and thereby function as a solar cell.

That would be awesome. Discovering a new phenomenon like this is always interesting, and has the potential to lead to new technologies. But you can see how casually now it is taken for granted that such discoveries will lead directly to specific applications. The valley of death awaits.

I am not trying to sound cynical, only realistic. We do have to keep trying to translate discoveries into applications, and most will fail. That’s OK. What the article discusses is that scientists need to have a realistic view of the hurdles in front of them. They need to, metaphorically, pack enough water so they don’t die of thirst  in the valley of death. This means getting enough funding. It also means not being overly optimistic, and understanding all the potential pitfalls of technology, such as the need for efficient mass production.

I would add that I would appreciate if the reporting on such advances dialed back the hype a bit and put things into a more realistic perspective.

They also discuss possible solutions, such as crowdfunding. Some scientists use crowdfunding websites to get them across the valley of death, when traditional funding sources dry up. I think this is a great option, but it comes with its own pitfalls.

The biggest, in my opinion, is that crowdfunding favors unrealistic hype. The poster child for this is the “solar freakin’ roadways” campaign. The hype was masterful, but didn’t address the many technological questions this idea raises.

Unrealistic but not impossible schemes also blend imperceptibly into rank pseudoscience. This is exactly why I think it is so important for science reporting to put new potential advances into a starkly realistic context. Plausibility and potential obstacles need to be given equal time to gee-whiz potential applications. Otherwise the public tends to become complacent about science-fictiony claims for the next technological advance.

Then, when someone comes along claiming to have an engine that runs on water perhaps the public will be a little less accepting and a little more critical.

Conclusion

The valley of death is essentially about translational research – taking basic science discoveries and translating them into practical applications. This is often the purview of private industry (more so than basic science). For any research, funding is the life’s blood and insufficient funding is death.

It is interesting to think about the system we have to support translational research, and what theoretical system would be the most efficient. Some would argue that the marketplace is the best tool for sorting out winners from losers in a Darwinian process, and this is a reasonable position. Darwinian processes, however, tend to be blind to the long term, and can be inefficient.

Top down processes can take the long view and can steer the ship, as it were. However, predicting specific winners and losers is probably too complex to be done by committee.

Crowdfunding adds a third democratic option. This option has its own strengths, but also may encourage hype and even fraud and pseudoscience.

Perhaps a blend of all options is optimal.